v0.8 — Apache-2.0 — replica fan-out shipped

Search. Auth.
Realtime. Replicas.
One binary.

kynetradb is a self-hostable backend engine — BM25 full-text search, HNSW vector similarity, auth, realtime channels, file storage, and read replicas — compiled into a single Rust binary that runs on any cloud, any VPS, or your own machine. No fleet of containers. No vendor. You own everything.

Deploy in 1 click Read the docs →

@kynetra/clientTypeScript SDK

14Cloud providers

1.07 msBM25 @ 100k

2.21 msVector @ 100k

~0.75sReplica lag

Measured, with the caveats attached

Numbers from the dated benchmark ledger committed alongside the engine. Each names its axis and its baseline. Two things you should know before reading any of them: the Caustic figure is from a synthetic corpus with real-embedding validation still pending, and the Disk per-kind index ships default-off because its variance ship-gate was not met.

Method, baselines and caveats →

0.13% SPANN resident vs raw f32

0.9950 Caustic recall@10 · 768-dim

3.0 ms Tenant read · 1M background

0.1% Write cost of the extra index

2,000 Raft simulation seeds

0.13% is resident memory as a fraction of raw f32, against the previous RAM HashMap (~125x less, byte-identical results). 3.0 ms is p50 tenant-scoped read latency at 1M background entities, versus 82.5 ms with the index off (~28x on Ram). 2,000 Raft seeds are an opt-in sweep, not the default test run, and are not a claim of distributed readiness — see reliability.

@kynetra/client — TypeScript SDK

A fully-typed TypeScript SDK with a fluent builder chain — from().select().eq().order() — covering queries, auth, storage, realtime channels, and edge functions under one import. Works in any TypeScript/JavaScript runtime: Node, Deno, Bun, browser.

quickstart.ts typescript

import { createClient } from '@kynetra/client'

const kdb = createClient('https://api.myapp.com', PUBLISHABLE_KEY)

// Query — fluent builder, fully typed
const { data, error } = await kdb
  .from('posts')
  .select('id, title, author(name)')
  .eq('published', true)
  .order('created_at', { ascending: false })
  .range(0, 19)

// Auth
const { data: { session } } = await kdb.auth.signInWithPassword({ email, password })
const { data: { user } }    = await kdb.auth.getUser()

// Storage
await kdb.storage.from('avatars').upload(`${user.id}.png`, file)
const { data: { publicUrl } } = kdb.storage.from('avatars').getPublicUrl(fileId)

// Realtime — change events + presence + broadcast
kdb.channel('live')
  .on('postgres_changes', { event: 'INSERT', table: 'messages' }, payload => {
    console.log('new message', payload.new)
  })
  .subscribe()

// Edge functions
await kdb.functions.invoke('send-welcome', { body: { userId: user.id } })

// Full-text + vector search
const { data: hits } = await kdb.from('docs').textSearch('body', 'rust async')
const nearest = await kdb.rpc('vector_search', { query: embedding, top_k: 10 })

Install

npm install @kynetra/client

ESM + CJS + TypeScript types included. No peer dependencies. Works with Next.js, Remix, SvelteKit, React Native, and Node scripts.

Connect

const kdb = createClient(
  process.env.KYNETRA_URL,
  process.env.KYNETRA_KEY
)

12 data-access integrations

Prisma and Drizzle shaped HTTP previews plus bridge starters for ten established ORM ecosystems. Every entry publishes its transaction, migration, and feature boundary.

Open the integration matrix →

Prisma

Node.js / TypeScript

Native HTTP preview preview

A Prisma-shaped repository for findMany, findUnique, create, update, delete, and upsert. It is not a Prisma driver adapter and does not run Prisma query-engine plans.

crud: supported scalar filters: supported transactions: unsupported vector search: raw-query

Drizzle

Node.js / TypeScript

Native HTTP preview preview

A Drizzle-shaped select builder with typed condition nodes and a bridge pgTable schema. It is not a drizzle-orm database driver.

crud: supported scalar filters: supported transactions: unsupported vector search: raw-query